Loading…

Generalized Eddington analytical model for azimuthally dependent radiance simulation in stratified media

A fast analytical radiative transfer model to account for propagation of unpolarized monochromatic radiation in random media with a plane-parallel geometry is presented. The model employs an Eddington-like approach combined with the delta phase-function transformation technique. The Eddington approx...

Full description

Saved in:
Bibliographic Details
Published in:Applied optics (2004) 2005-10, Vol.44 (28), p.6032-6048
Main Authors: Marzano, Frank S, Ferrauto, Giancarlo
Format: Article
Language:English
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A fast analytical radiative transfer model to account for propagation of unpolarized monochromatic radiation in random media with a plane-parallel geometry is presented. The model employs an Eddington-like approach combined with the delta phase-function transformation technique. The Eddington approximation is extended in a form that allows us to unfold the azimuthal dependence of the radiance field. A first-order scattering correction to the azimuth-dependent Eddington radiative model solution is also performed to improve the model accuracy for low-scattering media and flexibility with respect to use of explicit arbitrary phase functions. The first-order scattering-corrected solution, called the generalized Eddington radiative model (GERM), is systematically tested against a numerical multistream discrete ordinate model for backscattered radiance at the top of the medium. The typical mean accuracy of the GERM solution is generally better than 10% with a standard deviation of 20% for radiance calculations over a wide range of independent input optical parameters and observation angles. GERM errors are shown to be comparable with the errors due to an input parameter uncertainty of precise numerical models. The proposed model can be applied in a quite arbitrary random medium, and the results are appealing in all cases where speed, accuracy, and/or closed-form solutions are requested. Its potentials, limitations, and further extensions are discussed.
ISSN:1559-128X
DOI:10.1364/AO.44.006032